Research map / Probabilistic and scientific modeling / Variational and generative sampling
Neural Sampling in Hierarchical Exponential-family Energy-based Models
Xingsi Dong, Si Wu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use nonlinear statistic sigmoid(x) instead of linear x in HEE (ablation) | HEE (HEE-NL vs HEE-L) · 2D synthetic datasets and FashionMNIST | generation quality (qualitative) | Improved |
| marginal generation instead of joint generation (ablation) | HEE-NL-A · CIFAR10 unconditional | image quality (Inception score, FID) | Improved |
| marginal generation instead of joint generation (ablation) | HEE-NL · 2D synthetic datasets and FashionMNIST | convergence speed (qualitative) | Improved |
| add neural adaptation as momentum (SLD sampling instead of LS) | HEE-NL · inference sampling speed | sampling speed | Improved |
| increase adaptation strength m | HEE-NL-A · CIFAR10 inference | oscillation frequency and onset transients: 20-80 Hz | Improved |
| deeper layers of the network | HEE-NL-A · CIFAR10 linear SVM on latent responses | classification accuracy | Improved |
| joint generation with HEE-NL | HEE-NL · 2D synthetic datasets | mode coverage | Hurt |
Closest papers by meaning
- Energy-Based Transformers are Scalable Learners and Thinkers (ICLR 2026 · similarity 0.87)
- Brain-like Variational Inference (NeurIPS 2025 · similarity 0.87)
- eXponential FAmily Dynamical Systems (XFADS): Large-scale nonlinear Gaussian state-space modeling (NeurIPS 2024 · similarity 0.86)
- Learning Latent Space Hierarchical EBM Diffusion Models (ICML 2024 · similarity 0.86)
- Complex priors and flexible inference in recurrent circuits with dendritic nonlinearities (ICLR 2024 · similarity 0.86)
- Learning Energy-Based Models by Cooperative Diffusion Recovery Likelihood (ICLR 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
